It is not surprising that the decision by Governor Cuomo to give Amazon $1.8 billion in grants and refundable tax credits to come to New York City for half of their second headquarters generated controversy. Some have questioned the need to subsidize Amazon given New York’s labor pool advantages, and the amount of money given the company to lure them to New York. Others have questioned why New York would pay billions to encourage a business to locate in the part of New York State that is already doing well when the economy of much of the rest of New York is stagnant. Residents of the area expressed concerns about likely higher housing prices and additional congestion resulting from the project.
To be sure, the process used by Amazon to choose its second headquarters sites, aptly dubbed “the hunger games” was structured to create a bidding war between potential locations. And, there is no doubt that granting location incentives to businesses like Amazon has many downsides:
- It shifts the costs of existing government services to other taxpayers in unfair ways.
- It imposes additional subsidy costs on taxpayers who do not benefit.
- It creates an expectation by other businesses considering relocations or expansions that they can get the same kinds of subsidies that Amazon received.
New York’s offer to Amazon consisted of a $505 million capital grant toward the $3.686 billion project cost and up to $1.2 billion in refundable tax credits through the state’s Excelsior program. Virginia offered an incentive package including nearly $600 million in cash grants, plus $195 million in infrastructure improvements. Amazon’s project cost in Virginia is expected to be $2.5 billion.
But, the benefits of the Amazon project to New York are real – 25,000 jobs paying an average of $150,000 annually to employees. In fact, the Amazon project is by far the largest business attraction opportunity in memory. Even in a metropolitan area as large as New York’s, the economic impact is significant, adding $3.75 billion in payroll spending to the metropolitan area each year, with a total annual economic benefit to the state of about $7.5 billion. The state estimates that the project will generate $560 million annually. The cost of incentives would be paid back with additional tax revenue in about three years.
It is helpful to understand the challenges faced by state leaders in responding to Amazon’s second headquarters project. The process involved in the Amazon location decision differed from virtually all previous corporate location decisions because of Amazon’s decision to create a public bidding war between locations. That meant that every large city in the United States would put together an incentive proposal for Amazon HQ2 – 238 in all.
In most business attraction cases, companies work with site selection consultants that help them identify needs and wants. The consultants winnow down potential sites to a small number that receive serious consideration. From that point, negotiations between companies, consultants and government take place confidentially.
Amazon eventually announced publicly that it had reduced the list of locations being considered to 20. Once confidential discussions began between Amazon and state and local governments, information about competitive offers became harder to obtain. But there was reason to believe that a successful incentive offer would have to be substantial to be successful.
Decision-makers in state and city governments knew that New Jersey’s proposal would receive serious consideration (New Jersey offered $7 billion in incentives). Because much of New Jersey lies within the New York metropolitan area, it shares the same labor pool advantages offered by New York. And, as Amazon’s ultimate decision demonstrated, the Washington D. C. metropolitan area had many of the same advantages offered by New York City, including a large pool of technology workers (Amazon received an $8 billion proposal from Maryland).
When governments negotiate corporate location incentives, they face some inherent disadvantages because they have less information than the business. Government negotiators can estimate tax incentives available from competitive locations, but do not know how the company values them. State and local negotiators cannot be certain what discretionary incentives are being offered by competitors. Nor does government know how the company views the advantages and disadvantages of the sites that it is considering. Finally, they cannot know whether companies are telling the truth when they make representations about other offers that they have received.
The difference in sizes between New York’s and Virginia’s incentive packages probably reflects several factors. It’s likely that one reason New York provided more assistance is the fact that Amazon’s New York project will cost almost $1.2 billion more than the Virginia project. A second factor may be that New York advertises the availability of Excelsior tax credits. Because these benefits are visible to the public in New York’s business marketing materials, businesses considering New York locations expect to receive them as a matter of course. As a result, they are not a subject of negotiation.
Additionally, long-term tax credits like the Excelsior program do not use public dollars efficiently. Research shows that long-term tax incentives are heavily discounted by businesses considering new locations because benefits paid soon are heavily favored over those in future years. Reducing or eliminating reliance on tax credit programs like Excelsior could save the state money for future business attractions.
It could be argued that states like New York should refuse to engage in incentive wars like Amazon’s, but such a course of action would be difficult from a political perspective. Any governor or mayor who took his or her state out of competition for thousands of jobs would provide potential future opponents with campaign fodder.
From a public policy perspective, although there are significant negative tradeoffs associated with a public subsidy of the magnitude that was provided to Amazon, the reality is that 25,000 good-paying jobs are too many to give away to another location. Even with the very large public subsidy provided to Amazon, there is a substantial net economic and fiscal benefit to the State from securing the jobs for New York.
In the end, financial incentives alone did not determine the choices that Amazon made. In fact, Amazon passed up larger incentives from direct competitors to New York (New Jersey) and Virginia (Maryland). But, to conclude that New York or Virginia could have landed Amazon’s HQ2 without the use of incentives is wishful thinking.
 A recent Brookings Institution brief shows that the New York Metropolitan area has by far the largest number of workers in computer and mathematical occupations. https://www.brookings.edu/blog/the-avenue/2018/11/13/for-amazon-hq2-location-decision-was-about-talent-talent-talent/
 Estimates derived from: https://www.governor.ny.gov/news/governor-cuomo-and-mayor-de-blasio-announce-amazon-selects-long-island-city-new-corporate
  Timothy Bartik, 2009. “What Works in State Economic Development?” In Growing the State Economy: Evidence-Based Policy Options, 1st edition, Stephanie Eddy, and Karen Bogenschneider, eds. Madison, WI: University of Wisconsin, pp. 19.
Well stated, but as Twitter commentators have said, there is no universe in which Amazon does not place a significant branch presence in NYC regardless of incentives, so I’m not sure the counterfactual is losing all 25,000. In any smaller jurisdiction struggling for recognition, the decision to offer subsidy (even a rich one) is easy to understand. Here it’s harder, though you lay out the reasons it’s still tempting. I’m starting to think it’s a lose-lose for Bill de Blasio. If he “wins” over Council objection, he takes it from the left. Hard. If he loses, he looks inept (ahem) and lame-duckish. His would-be successors are going to have to choose sides on this, and it won’t be pretty, especially with all the tech companies coming under enormous pressure for socially amoral decision-making.
Good points, David. Of course Amazon is already in NYC with several thousand jobs, but Jersey City or Newark were viable alternatives for the HQ2 jobs, and in a split decision, more of the jobs could have gone to Alexandria, VA. So, I just don’t think NYC was the only viable location for the Amazon HQ2 location. For that reason, I think that the adminstration’s willingness to offer incentives was justified, though I have reservations about the value of the Excelsior program tax credits.
Hope all is well. I enjoyed reading your article about the Amazon deal but am not sure I understand your comment about the Excelsior Jobs Program. This Program is a negotiated tax credit that is fully refundable so it is the equivalent of cash. It is, in essence, a grant disguised as a tax credit, and, the award amount is negotiated. Few Program participants are ever offered the maximum tax credits allowed under the law. So while companies may expect the maximum credits as if it were an “as of right” benefit, there is no guarantee of receiving that amount. This feature of the Program was culture shock to many companies and economic practioners, including ESD project originators, when the Program was first launched eight years ago. Over time, everyone adjusted to the new world order.
While true it is usually paid out over a longer period than a grant, this feature makes it better for taxpayers because each year’s payout only happens after job performance is verified and performance is measured against a statewide base employment. Grants often only consider the incremental jobs at the project location when performance evaluations are done so companies can lose jobs in one part of the State and still be in compliance with the terms of the grant. In other words, under the Excelsior Jobs Program all existing and new jobs must be maintained for a longer period of time in order to receive the full payout of the offer.
Another feature that makes the Excelsior tax credits a better deal for taxpayers is that the disbursement in any particular year must, by law, be prorated if job performance is below the commitment. If less than 75 percent of a job goal is met, no disbursement can be made for that year and the company loses that year’s benefits. By comparison, with a grant there is generally a 15 percent allowance given so a company can fall short of the job goal in a particular year and still receive the full disbursement in the schedule. If the job shortfall exceeds the 15 percent allowance then ESD can negotiate a recapture or a revised payout. In the Amazon deal, any disbursements recaptured or withheld will be paid out if 100 percent of the job goal is met in the 5th, 10th or 15th project year.
So I guess if your point about Excelsior was that you question the tax benefits because of the time value of money, that may be true in many cases but Excelsior better protects taxpayers. In the Amazon deal, there is no time value of money difference because the tax credits are paid over 10 years and the grant 15 years. However, Amazon has more assurance of receiving the grant funds, even if there are shortfalls in job performance. And, it is not clear whether Amazon must maintain all existing 4,747 employees elsewhere in the State when job performance is measured. In Excelsior they must do so by law.
I would argue that the Excelsior Jobs Program is valued by companies as evidenced by the number of ESD deals done with Excelsior as the only incentive and it being the biggest part of the Amazon deal but New York taxpayers are better protected as well.
Thanks, Randy – this is useful additional information.